S-XPS 1645 AC Power Throttle Issue Investigation

How are the performance results with the 210W?

 

I'll let you know when it gets delivered.

 

News from the French front :

The case reached the french tech press : http://www.presence-pc.com/actualite/Dell-E6400-E6500-1645-37365/#xtor=RSS-11.

They have contacted Dell FR who officialy stated that the problem has been investigated and solved by the release of BIOS update. Depending on your system (Latitude or XPS), the BIOS may not be available yet.


Any feedback on this BIOS update yet ? One of our french user tested it on a Code 2 Duo system and noticed that throttling message in his performance monitor has disapeared since he updated his BIOS and that the system is way more reactive.

 

I just got the adapter in, and it does seem to be coupled with some bios issues as well. Although i cant kill the 130w adapter like i could the 90w when under full load, so thats good news.

I will say the 130w does seem to run better, but throttling is still taking place. It would seem whenever 2 things causing load it clocks down the cpu. Im busy til later tonight ill try to get actuall results. But to say the pa-4e i got stays cool which is awesome. And it is the slim version which is just slightly less then twice as thick as the 90w adapter.

edit: cool helmut heres the google translated link for us english readers

 

anyone know the cheapest place to get the 210W?

preferably for canada

and atlstang does the 130 have the neeto little light up end on the power cord?

oh and the model #

thaaaanks

 

The bios are :
A18 for Latitude E6500
A19 for Latitude E6400
Don't seems to be available for XPS 16 yet.

It is yet to be confirmed that the Latitude and XPS issues are related.

 

lol I called about my lack of windows7 and while I was talking to the guy at technical support I said that I have a problem with my sound and fps when playing some new games on high. the guy was gone for a minute and then said that this is a known issue and they are researching it and will likely release an update for this issue soon. He said that it is not a problem to worry about or an ac adapter issue. He said that several times even though I never questioned him. I haven't actually noticed this on my comp because I don't have any games installed yet. so I didn't pursue it.

 

Probably the BIOS update is for the latitude family, as they are less power demanding than the xps 1645 and the 90 watts is not enough.

We have provided enough evidence that it is a power problem which may be:
Power supply, BIOS, or hardware problems.
The normal interpretation and the evidence we have seen so far is that it is likely to be the power supply as all the other i7 laptops come with at least 120W even though the XPS is more power consuming. if CurtDawg77 finds the 210W adapter working fine, then case is close regardless of whatever BS the DELL technicians are going to say.
BIOS and/or hardware can be factors, but they seem not to be the main issue unless their genius engineer decided to make the throttling during the Battery as well in their BIOS update and hence no performance drop from changing from battery to AC
I don't think that a capacitor problem or a BIOS problem will result in a 25-40watt power requirement increase.

 

khal can you explain your testing you did. It was prime95 and crysis gpu benchmark? any modification to the crysis test? and what programs where you monitoring with. I got the 130w in and still see downclocking when should be full load, want to try your tests out and see whats the difference.

 

so - on that note, for those of us who want to be extra safe and want more than the 130W; lets scour the intranets for the best deals on the 210W

i dont see it on trianglecomputers

the 210W was for the alienware yes?

 

We will see if a BIOS update for the XPS 16 shows up. But I think so to, the issues don't look to be related.
I would be surprised that an i7 + HD4670 requires less than 90watt considering blue-ray and rgbled.

 

Prime95 and crysis gpu benchmark in a window mode (there is a software that you can set the different parameters Link) run the 32bit version not the 64bit and when it finishes you will receive an error that it is trying to close crysis go ahead and accept it, Batterymon to monitor battery discharge, Rivatuner to monitor the GPU clock, RealTemp i7 to monitor the CPU temp and its multipliers.

Both cases under max performance power plan with powerplay set at maximum performance.

Under battery multipliers changed between 10-12x discharge reached 105watts
under AC multipliers stayed at 7x and No charging at all (and this explain the error some people get when trying to pull and/or plug the charger at load).

In all cases the CPU never reached more than 75-77c.

If you get the full 13x then 130w is more than enough.
If you get 11-12x, then it is ok.
if you get 9x-10x then it is 130 is better than 90 but not enough though.
if you get 7x, then it is not related to the charger
Waiting for your results
thanks

 

I got 1x sometimes on 90 watt.

 

Mine would stay around 8.5x-11x when using the 130w I have under max load with full brightness, it has no problem with screen brightness on min though (13x non stop). It’s really surprising how much wattage the RGBLED uses. kudos to anyone that can find the specs on the RGBLED.

BTW- the speakers alone on this system consume 7W.

 

didn't happen to me, the only difference between our configuration is the WLED vs the RGBLED.
So it may be a bios error in managing the colors in the RGBLED, do you get the same on an external monitor?

 

From what i have seen so far, It looks like the 130w is at border or just fine for the WLED configuration but less than what the RGBLED configuration needs. can you measure the performance on an external monitor.

 

guys does having stuff plugged into the usb ports charging from it etc. weigh on the adapter too?

it would wouldn't it?

some of us have cooling pads and iPods and the like connected

anyone try any tests with that stuff? or do iPods and that hardly draw any power?

 

this is all i have found so far....

http://www.hardforum.com/showthread.php?p=1034825325

 

i dunno folks.. i'm only guessing, but the 130W adapter has over 44% greater performance than the 90W and dell's (one of the best laptop manufacturers worldwide) engineers thought 90W would be enough. so even if it's true that the 90W adapter's not enough for all systems, if the 130W wouldn't be enough they must have really badly screwed up something.. +40W is a huge extra power to 90W.

 

This leads me to the belief that there is something wrong in the BIOS regarding the RGBLED in conjunction with the under powered charger, it consumes nearly 35% more than the WLED not enough to cause the reported huge performance drop.

 

No I always get excellent performance on the external monitor (with laptop display turned off), I actually perfer to play games on there so it's not a big deal. Note that it doesn't stay at 1x, its constantly jumping up and down. I read somewhere that Core i7s throttle 100 times per second, so the monitoring tools dont give an accurate picture.

 

It shouldn't hit 1x anyways. This is for sure a BIOS problem for the RGBLED. In addition to the under powered charger i should say.

 

Yeah the Dell reps insisted on me waiting until their BIOS update is available. They said it should be available around the 15th, they're working with Intel on it right now. I couldn't really argue, a BIOS update could change the throttling behavior although I'm still skeptical that a 90 watt adapter is sufficient.

 

I dont have time to do a thorough test til later tonight.. but i tried it out just to see. What i found was battery i can get 12x everything going. But didnt matter the 130w or the 90w it still throttled to 7x(even if i told prime95 to do just one process). Will note though the 130w did seem to do better again as in smoother, but i need to look into it more thoroughly as thats just what it seems. rivatuner was giving me issues, will have to sort later.

I agree i dont think 90w is enough as i can kill it under full load. The 130w is much cooler then the 90w but also is bigger. A lot of wierdness is going on for sure, as you cant just unplug and plug in a new adapter. Resetting the test is required to get it out of the 7x state for battery etc. I dunno theres issues for sure. All i can say from the looks of it the 130w does better, but still having issues.

 

Yeah i forgot to mention that this is a problem to the RGBLED in addition to the adapter, i added it to my comment.

 

Please confirm as this contradicts our claim, this will be a system problem (BIOS and/or hardware) rather than being a charger problem.

 

1x is impossible to obtain on an i7. From what the developer of RealTemp tells me, the processor physically can't run at that speed: what were you monitoring with? Whatever it was, that can't be right.

~Ibrahim~

P.S. If the CPU is stuck at 7x, you should restart the system. Also disable EIST, reboot to Windows, reboot to BIOS and enable, reboot to Windows.

 

I know i really want to do it now, but that will have to wait til later. I have classes til later tonight. ill make sure to do so.

 

That is what i thought also.

 

o no.... will the BIOS update business delay my delivery? cuz i have to have it by the 15th... then i won't be in the states for like 8 months...

 

Well, it didn't hold up mine - shipped today.

 

There definitely has to be a BIOS issue here, I really hope there is an updated BIOS very soon here for this system... granted the 130W should be enough, but until someone can run a device to check wattage draw on a 130W adaptor it’s hard to say for sure.

i dont have my laptop with me at the moment, but can someone do this and return what it says?

http://www.laptopshop.co.uk/laptop-notebook-ac-adapter/tip-packs.htm

 

argh... fingers crossed... we have about the same specs except for the HDD (mine is 500GB SATA). Would you mind testing your machine after you get it? and note if they are really giving out 130W adapters for new 1645 laptops now...

 

nobody answered me about the USB power draw

is that going to be an issue too?

 

Yes, USB Devices drain power from the AC Adaptor, some more than others.

 

i just purchaced a Kill-A-Watt 4400... i should get it next week.

 

According to this website, RGBLED only uses 5 more watts.

 

well fudge then the 90W cant be enough

 

me too, here are my specs

It provided 5300 kilowatts (4400 kilowatts for hot pressurized gas and 900 kilowatts for electricity)'


lol

 

just called Dell... apparently the technician who was talking to me wasn't aware of this problem, but he was nice enough to talk to his supervisor, who knew wut's going on. And according to his supervisor, Dell is not giving out any 130W adapters to fix the problem, but the engineering department is currently working on it. So i asked why ppl are getting 130W adapters, he said it's because ppl are insisting...

I don't have time to go all the way up the food chain to talk to their super-super-supervisor, nor do i have enough knowledge to argue with them. Another disadvantage I have now is that I don't have the system with me yet, so can't give out concrete examples from my side. For ppl who did get the 130W adapter, what did you say to Dell?

 

Would be useful if anyone who speaks to Dell Customer support, let us know which country you are in USA, UK, France...etc. as it seems their response differ from one region to the other.

 

yeah thats another issue

just because dell UK recognizes this doesnt mean Canada does

someones gotta talk to the corporate office or something

and whats worse - i think i've got Temporomandibular joint disorder

 

Isn't all the technical support calls managed by DELL in INDIA?

 

I contacted Dell US. chatted online once, and called once, both were saying it's not adapter problem.

 

possibly

everyone i've talked to online was indian

but they were on eastern time so nawww

 

see now that makes me more worried that im gonna get a release model 1645 with faulty hardware - then they'll fix the bloody things

probably is the mobo mishandling the power

or some witchcraft

 

however the online guy said i will be shipped with the newly configured motherboard. sounds comforting...

 

best.....answer.....yet.....

 

when did you place your order, whens it expected to arrive, and what country

i assume they'll be doing this for all currently placed orders

dells pullin a 360 on us

 

I ordered it on US Dell website on Nov. 14, estimate delivery date is Dec. 9. I specifically asked them if the order is gonna be delayed by this issue, and i got the answer no.